docker/sbx-releases

Docker Sandboxes run AI coding agents in isolated environments on your machine or on Docker-managed cloud infrastructure

What it solves

Docker Sandboxes provide safe, isolated environments for AI agents to operate autonomously. This prevents agents from accidentally or intentionally putting the host machine or sensitive data at risk when executing code or accessing files.

How it works

It uses Docker-native isolation and containerization principles to create sandboxes with controlled access to the filesystem, network, and tools. It is vendor-neutral, meaning it can be integrated with various AI models and tools like Claude Code, Codex, and Gemini CLI.

Who it’s for

Developers building or using AI agents that need to execute tasks autonomously on a local machine without compromising security.

Highlights

  • YOLO mode: Agents can work without needing constant permission requests.
  • Private Docker daemon: Includes a dedicated daemon for running test containers.
  • Control mechanisms: Fine-grained access controls for both the network and the host-to-sandbox filesystem.
  • Broad compatibility: Works with multiple agentic tools including OpenCode, Gemini CLI, and Claude Code.

Related

  • Project
    anomalyco/browser-control

    Browser Control is an open‑source Node.js driver that lets AI agents or CLI users run Playwright code inside their own Chromium browser (with all logged‑in sessions and extensions). It works via a local relay and a browser extension, supports persistent sessions, inspection helpers, human hand‑offs, video/ HAR recording, and a TypeScript Effect client for secret‑safe API calls. Intended for trusted local use only.

  • Project
    tensorflow/tflite-support

    A toolkit for developing and deploying TFLite models to mobile devices, providing tools for metadata management, automatic API generation, and optimized task libraries.

  • Project
    nebius/soperator

    A Kubernetes operator that automates the deployment and management of Slurm clusters for AI training and HPC, featuring automated health checks and GPU optimization.

  • Project
    stan-dev/bayesplot

    An R package providing a library of plotting functions for visualizing posterior draws, MCMC diagnostics, and predictive checks for Bayesian models.